CN112634254A - Insulator defect detection method and related device - Google Patents
Insulator defect detection method and related device Download PDFInfo
- Publication number
- CN112634254A CN112634254A CN202011610362.4A CN202011610362A CN112634254A CN 112634254 A CN112634254 A CN 112634254A CN 202011610362 A CN202011610362 A CN 202011610362A CN 112634254 A CN112634254 A CN 112634254A
- Authority
- CN
- China
- Prior art keywords
- insulator
- image
- target
- detected
- defect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000012212 insulator Substances 0.000 title claims abstract description 513
- 230000007547 defect Effects 0.000 title claims abstract description 192
- 238000001514 detection method Methods 0.000 title claims abstract description 144
- 238000000034 method Methods 0.000 claims abstract description 62
- 230000002950 deficient Effects 0.000 claims description 82
- 238000003062 neural network model Methods 0.000 claims description 56
- 238000013528 artificial neural network Methods 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 13
- 238000004088 simulation Methods 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 238000007689 inspection Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the application provides an insulator defect detection method and a related device, wherein the method comprises the following steps: acquiring an image to be detected; under the condition that the image to be detected comprises the insulator, segmenting a target insulator image from the image to be detected; and detecting the defects of the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects. The embodiment of the application is favorable for improving the efficiency of insulator defect detection.
Description
Technical Field
The application relates to the technical field of image detection, in particular to an insulator defect detection method and a related device.
Background
The power transmission facilities of the national power grid and the running route of the Chinese railway are important infrastructure for the society to live and develop, so that the routing inspection work of the power transmission line is particularly important in the aspect of guaranteeing national life production and industrial development. At present, a method for identifying defects of insulators in an inspection line for power transmission is as follows: the inspection vehicle collects images of the whole power transmission line and uploads the images at a certain time frequency, monitoring personnel observe the whole images through human eyes, the position of the insulator is located, whether the insulator is defective or not is checked, and the efficiency is low when the insulator defect detection is carried out in a manual inspection mode.
Disclosure of Invention
The embodiment of the application provides an insulator defect detection method and a related device, which are beneficial to improving the efficiency of insulator defect detection.
A first aspect of an embodiment of the present application provides a method for detecting an insulator defect, where the method includes:
acquiring an image to be detected;
under the condition that the image to be detected comprises the insulator, segmenting a target insulator image from the image to be detected;
and detecting the defects of the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects.
The method comprises the steps of detecting an image to be detected, segmenting the image to be detected if the image to be detected has the insulator, obtaining a target insulator image, further detecting the defect of the insulator in the image based on the target insulator image, and determining whether the insulator has the defect.
With reference to the first aspect, in one possible implementation manner, the dividing the target insulator image from the image to be detected includes:
acquiring an insulator image from an image to be detected;
determining the insulator image as a target insulator image under the condition that the size of the insulator image is larger than or equal to a preset size;
and (4) segmenting the insulator image from the image to be detected to obtain a target insulator image.
In this example, the image size of the insulator image is greater than or equal to the image with the preset size, and the image is determined as the target insulator image, so that the image which is not suitable for insulator defect detection can be filtered, and the efficiency of defect detection on the insulator is further improved.
With reference to the first aspect, in one possible implementation manner, the method further includes:
determining the insulator in the insulator image as an interference insulator under the condition that the size of the insulator image is smaller than the preset size;
and filtering the image to be detected, and not executing the operation of segmenting the insulator image from the image to be detected.
In this example, the insulator in the insulator image with the size smaller than the preset size is determined as the interference insulator, the image to be detected containing the interference insulator is filtered, and the segmentation is not executed, so that the efficiency of defect detection on the insulator is improved.
With reference to the first aspect, in one possible implementation manner, performing defect detection on a target insulator image to determine whether an insulator in the target insulator image has a defect includes:
performing defect detection on the target insulator image to obtain at least one first detection frame;
and determining whether the insulator in the target insulator image has a defect or not according to the confidence coefficient of the at least one first detection frame.
With reference to the first aspect, in one possible implementation manner, determining whether an insulator in a target insulator image has a defect according to the confidence of at least one first detection frame includes:
determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
and determining that the insulator in the target insulator image has no defect if the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is less than the threshold value.
In this example, whether the insulator in the target insulator image has a defect is determined by the defect probability corresponding to the detection frame with the highest confidence, so that the accuracy in judging the insulator defect can be improved.
With reference to the first aspect, in one possible implementation manner, the defect detection on the target insulator image is performed by a first neural network model; the method further comprises the following steps:
acquiring a false positive rate of the first neural network model and acquiring a recall rate of the first neural network model;
a threshold is determined based on the false positive rate and the recall rate.
In this example, the threshold is determined according to the false positive rate and the recall rate, and then the threshold can be determined in a feedback mode, so that the judgment of whether the insulator has defects is corrected, and the accuracy in the judgment is improved.
With reference to the first aspect, in one possible implementation manner, the first neural network model is obtained by training through the following steps:
acquiring a first sample set; the first sample set includes a mask of a first defective insulator image, label data for the first defective insulator image, a second defective insulator image, and a second defective insulator image;
inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection so as to respectively and correspondingly obtain at least one second detection frame;
obtaining a defect detection result according to the confidence of the at least one second detection frame;
and adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
In this example, when the first defective insulator image collected in the inspection line is insufficient, the second defective insulator image and the mask thereof are used to expand the first sample set, and the expanded first sample set is used to train the first neural network model, which is beneficial to improving the performance of the first neural network model.
With reference to the first aspect, in one possible implementation manner, before obtaining the first sample set, the method further includes:
acquiring attribute information and background information of a target insulator;
and generating masks of a preset number of the second defective insulator images and the second defective insulator images by a 3D simulation technology by using the attribute information and the background information.
In this example, the predetermined number of second defective insulator images is determined by a 3D simulation technique, and then the first sample set may be expanded to add training data.
With reference to the first aspect, in one possible implementation manner, the defect type of the insulator in the second defective insulator image includes at least one of the following: insulator breakage and insulator flashover.
With reference to the first aspect, in one possible implementation manner, after the image to be detected is acquired, the method further includes:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise the first defect insulator image.
A second aspect of the embodiments of the present application provides an insulator defect detecting apparatus, which includes:
the first acquisition unit is used for acquiring an image to be detected;
the second acquisition unit is used for segmenting a target insulator image from the image to be detected under the condition that the image to be detected comprises the insulator;
and the determining unit is used for detecting the defects of the insulators in the target insulator image so as to determine whether the insulators in the target insulator image have defects.
With reference to the second aspect, in a possible implementation manner, in terms of segmenting the target insulator image from the image to be detected, the first obtaining unit is specifically configured to:
acquiring an insulator image from an image to be detected;
determining the insulator image as a target insulator image under the condition that the size of the insulator image is larger than or equal to a preset size;
and (4) segmenting the insulator image from the image to be detected to obtain a target insulator image.
With reference to the second aspect, in one possible implementation manner, the second obtaining unit is further configured to:
determining the insulator in the insulator image as an interference insulator under the condition that the size of the insulator image is smaller than the preset size;
and filtering the image to be detected, and not executing the operation of segmenting the insulator image from the image to be detected.
With reference to the second aspect, in a possible implementation manner, in detecting a defect of the target insulator image to determine whether the insulator in the target insulator image has a defect, the second obtaining unit is specifically configured to:
performing defect detection on the target insulator image to obtain at least one first detection frame;
and determining whether the insulator in the target insulator image has a defect or not according to the confidence coefficient of the at least one first detection frame.
With reference to the second aspect, in a possible implementation manner, in terms of determining whether an insulator in a target insulator image has a defect according to the confidence of at least one first detection frame, the second obtaining unit is specifically configured to:
determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
determining that there is no defect in the insulator in the target insulator image if the maximum confidence in the confidence of the at least one first detection box is less than the threshold.
With reference to the second aspect, in a possible implementation manner, the apparatus further includes a third obtaining unit, where the third obtaining unit is configured to:
acquiring a false positive rate of the first neural network model and acquiring a recall rate of the first neural network model;
a threshold is determined based on the false positive rate and the recall rate.
With reference to the second aspect, in one possible implementation manner, the third obtaining unit is further configured to:
acquiring a first sample set; the first sample set includes a mask of a first defective insulator image, label data for the first defective insulator image, a second defective insulator image, and a second defective insulator image;
inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection to obtain at least one second detection frame of each image in the first defective insulator image and the second defective insulator image;
obtaining a defect detection result according to the confidence of at least one second detection frame;
and adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
With reference to the second aspect, in one possible implementation manner, the third obtaining unit is further configured to:
acquiring attribute information and background information of a target insulator;
and generating a preset number of second defect insulator images and masks of the second defect insulator images by a 3D simulation technology by adopting the attribute information and the background information.
With reference to the second aspect, in one possible implementation manner, the defect type of the insulator in the second defective insulator image includes at least one of: insulator breakage and insulator flashover.
With reference to the second aspect, in one possible implementation manner, the first obtaining unit is further configured to:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise first defect insulator images.
A third aspect of embodiments of the present application provides an electronic device, including a processor, an input device, an output device, and a memory, where the processor, the input device, the output device, and the memory are connected to each other, where the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to perform some or all of the steps described in the first aspect of embodiments of the present application.
A fourth aspect of embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program for electronic data exchange, where the computer program makes a computer perform part or all of the steps as described in the first aspect of embodiments of the present application.
A fifth aspect of embodiments of the present application provides a computer program product, wherein the computer program product comprises a non-transitory computer readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps as described in the first aspect of embodiments of the present application. The computer program product may be a software installation package.
These and other aspects of the present application will be more readily apparent from the following description of the embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1A is a schematic diagram of a method for detecting an insulator defect according to an embodiment of the present disclosure;
FIG. 1B is a schematic diagram of a large graph and a small graph according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a method for detecting insulator defects according to an embodiment of the present disclosure;
fig. 3A is a schematic view of a polygonal frame of an insulator defect according to an embodiment of the present disclosure;
FIG. 3B is a schematic diagram of a defect target insulator image according to an embodiment of the present disclosure;
fig. 4 is a schematic flowchart of another insulator defect detection method according to an embodiment of the present disclosure;
fig. 5 is a schematic flowchart of another insulator defect detection method according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of an insulator defect detecting apparatus according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of another insulator defect detecting device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first," "second," and the like in the description and claims of the present application and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
In order to better understand the insulator defect detection method provided by the embodiment of the present application, first, a brief description is given below to the insulator defect detection method. Referring to fig. 1A, fig. 1A is a schematic view illustrating a method for detecting an insulator defect according to an embodiment of the present disclosure. As shown in fig. 1A, an image to be detected may be obtained by a camera or the like, and the image to be detected may be an image of the power transmission line, where an insulator may or may not be present in the image. The image to be detected can be detected through the second neural network model to judge whether insulators exist in the image to be detected or not, when the insulators exist in the image to be detected, a target insulator image is segmented from the image to be detected, the target insulator image can be understood as a small image comprising the insulators, the image to be detected is a large image comprising the insulators, and the large image and the small image can be shown in fig. 1B. And carrying out defect detection on the target insulator image by using the first neural network model so as to determine whether the insulator has defects. The first neural network model and the second neural network model are pre-trained detection models, the second neural network model is used for insulator detection, and the first neural network model is used for defect detection of the insulator. Therefore, compared with the prior art, whether the insulator has defects or not is detected in an artificial mode, the position of the insulator can be rapidly located through the detection model, whether the insulator has defects or not is detected, and the efficiency in defect detection is improved.
Referring to fig. 2, fig. 2 is a schematic flow chart illustrating a method for detecting insulator defects according to an embodiment of the present disclosure. As shown in fig. 2, the method includes:
201. and acquiring an image to be detected.
In this application embodiment, can obtain through camera, camera device etc. and wait to detect the image, camera etc. can set up in power transmission line's fixed position, for example, wire pole, wire tower etc. for example set up again on being close to the fixing device of wire pole, wire tower etc. can also set up on the circuit patrols and examines the car.
Since defects of the insulator may be caused by external effects during use, it is possible to acquire an image to be detected by setting an acquisition time interval according to which. Of course, the image to be detected may also be an input image, and the image may specifically be an image taken by a manager, and the like, and is not limited in particular here.
202. And in response to the insulator being identified in the image to be detected, segmenting a target insulator image from the image to be detected.
In the embodiment of the application, a pre-trained second neural network model can be adopted to perform target detection on an image to be detected, if the second neural network model has output, it is determined that an insulator exists in the image to be detected, and if the second neural network model has no output, it is determined that the insulator does not exist in the image to be detected. The output of the second neural network model may be an image including an insulator, or may be an image not including an insulator, and the image not including an insulator may be understood as an insulator in the image to be detected as an interfering insulator or an interfering object.
The method for segmenting the target insulator image from the image to be detected can be used for determining the target insulator image according to the size of a target detection frame in the output of the second neural network model, wherein the target detection frame is used for detecting the insulator. The target insulator image includes an image of an area framed by the target detection frame, and specifically, the target insulator image is an area framed by the target detection frame.
203. And detecting the defects of the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects.
In the embodiment of the application, a pre-trained first neural network model can be used for detecting the defects of the target insulator image, and whether the insulator has the defects or not is determined according to the defect probability of at least one detection frame during the detection of the insulator, wherein the detection frame is used for detecting the defects. For example, when the defect probability is greater than or equal to a certain threshold, it may be determined that a defect exists; and determining that no defect exists when the defect probability is lower than the certain threshold. Optionally, in the case that the insulator has a defect, the defect type may be further output.
It can be seen that, in this example, the image to be detected is detected through the second neural network, if it is detected that an insulator exists in the image to be detected, the image to be detected is segmented to obtain a target insulator image, the insulator in the target insulator image is further detected for defects by using the first neural network model, and whether the insulator has defects is determined.
In one possible embodiment, a possible method for segmenting a target insulator image from an image to be detected includes:
a1, acquiring an insulator image from an image to be detected;
a2, determining the insulator image as a target insulator image when the size of the insulator image is larger than or equal to a preset size;
and A3, segmenting the insulator image from the image to be detected to obtain a target insulator image.
In the embodiment of the present application, the insulator image may be understood as an image of an area where the target detection frame is output by the second neural network model, and the insulator image may be understood as a small graph.
The preset size may be set by empirical values or historical data, and one possible preset size is: 300 pixels by 300 pixels. After the insulator image is determined as the target insulator image, the insulator image can be cut from the image to be detected to obtain the cut individual target insulator image.
In this example, the image size of the insulator image is greater than or equal to the image with the preset size, and the image is determined as the target insulator image, so that the image which is not suitable for insulator defect detection can be filtered, and the efficiency of defect detection on the insulator is further improved.
In a possible embodiment, another processing may be performed when the size of the insulator image is smaller than the preset size, specifically as follows:
b1, determining the insulators in the insulator image as interference insulators under the condition that the size of the insulator image is smaller than the preset size;
b2, filtering the image to be detected, and not executing the operation of dividing the insulator image from the image to be detected.
In a possible embodiment, further processing may be performed when no insulator is identified in the image to be detected, specifically as follows:
and filtering the image to be detected under the condition that the insulator is not identified in the image to be detected.
In this embodiment of the application, the interference insulator may be, for example, an insulator on a cantilever of a contact network. The insulator in the insulator image with the size smaller than the preset size is judged to be the interference insulator, the image to be detected containing the interference insulator and the image without identifying the insulator are filtered, the segmentation is not executed, the efficiency of defect detection on the insulator is improved, and the extra calculation resource overhead is reduced.
In one possible embodiment, a possible method for performing defect detection on a target insulator image to determine whether an insulator in the target insulator image has a defect includes:
c1, carrying out defect detection on the target insulator image to obtain at least one first detection frame;
and C2, determining whether the insulator in the target insulator image has defects according to the confidence of the at least one first detection frame.
In a specific embodiment of the present application, the confidence of the first detection box may be determined through the first neural network model. And determining whether the insulator has defects or not according to the maximum value of the confidence degrees corresponding to the at least one first detection frame.
In this example, whether the insulator has a defect is determined according to the confidence of the at least one first detection frame, so that the accuracy of determining whether the insulator has a defect can be improved.
In one possible embodiment, a possible method for determining whether an insulator in a target insulator image has a defect according to the confidence of at least one first detection box includes:
d1, determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient in the confidence coefficients of the at least one first detection frame is larger than or equal to a threshold value;
and D2, determining that the insulator in the target insulator image has no defect under the condition that the maximum confidence coefficient in the confidence coefficients of the at least one first detection frame is smaller than the threshold value.
In the embodiment of the present application, the threshold is set by an empirical value or historical data. Here, the maximum confidence may be understood as a probability that the insulator has a defect, that is, the probability is greater than or equal to a threshold, and if the probability is less than the threshold, it is determined that the insulator has no defect.
In a possible embodiment, the present application further provides a method for determining a threshold, where the method includes:
e1, acquiring the false positive rate of the first neural network model and acquiring the recall rate of the first neural network model;
e2, determining a threshold value according to the false positive rate and the recall rate.
In the embodiment of the present application, the method for obtaining the false positive rate may be: obtaining a first number of negative samples detected as positive samples in a sample set of a first neural network model and a second number of correctly identified negative samples; and determining the false positive rate according to the first quantity and the second quantity. The false positive rate can be determined by a method shown in the following formula:
wherein FPR is the false positive rate, FP is the first quantity, TN is the second quantity, and N is the sum of the first quantity and the second quantity.
In an embodiment of the present application, the method for obtaining the recall ratio may be: obtaining a third number of correctly identified positive samples in the sample set of the first neural network model and obtaining a fourth number of positive samples detected as negative samples; and determining the recall rate according to the third quantity and the fourth quantity.
Specifically, the recall rate can be determined by the method shown in the following formula:
wherein RECALL is RECALL, TP is a third quantity, FN is a fourth quantity, and P is the sum of the third quantity and the fourth quantity.
When the threshold is determined according to the false positive rate and the recall rate, the specific false positive rate and the recall rate can be set according to the actual requirement, the threshold is determined according to the false positive rate and the recall rate, and the threshold can be determined through a preset determination relationship. The actual demand may be, for example, the number of defects detected and labor cost savings.
In a possible embodiment, the first neural network may be further trained through the sample to obtain a first neural network model, and the method specifically may include:
f1, acquiring a first sample set; the first sample set comprises a mask of a first defective insulator image, label data of the first defective insulator image, a second defective insulator image, and the second defective insulator image;
f2, inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection so as to respectively and correspondingly obtain at least one second detection frame;
f3, obtaining a defect detection result according to the confidence coefficient of the at least one second detection frame;
f4, adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
In this embodiment of the application, the sample image in the first sample set including the defective insulator may be an image in an existing data set, for example, a first defective insulator image acquired in an inspection line, or may be a sample image generated by a sample generation method, for example, a second defective insulator image generated by using a 3D simulation technique, where the sample image in the first sample set includes the first defective insulator image and the second defective insulator image, and the label data of the first defective insulator image and the mask of the second defective insulator image. Since the defects of the insulator are often small, the marking data is pixel-level, such as the polygonal frame shown in fig. 3A. The defect detection result may be a second detection frame with the highest confidence level and greater than or equal to the threshold, and may further include a defect type corresponding to the second detection frame.
The Mask-RCNN framework can be used as a first neural network for training, and the Mask-RCNN framework uses strong pixel-level supervision on the frame of the positive sample, so that the detection performance of the positive sample can be improved, namely the detection rate of the positive sample is improved.
In one possible embodiment, because the number of samples with defective insulators is small in an existing data set, such as an image data set collected in a routing inspection line, for example, in 20000 images with insulators in the existing data set, only less than 50 images including defective insulators are needed to be extended, a possible method for extending the samples includes:
g1, acquiring attribute information and background information of the target insulator;
g2, generating masks of the second defect insulator image and the second defect insulator image in preset quantity by a 3D simulation technology by adopting the attribute information and the background information.
In this embodiment of the application, the target insulator may be an insulator commonly used in a railway contact system, and attribute information of the target insulator may be obtained through query, for example, attribute information such as the type, size, and material of the commonly used insulator may be queried in a power supply system of the railway contact system. Of course, the attribute information of the target insulator may be obtained in other manners, for example, in a manner specified by a user, or in a manner of querying from the internet or a server.
The background information may be understood as background information of the target insulator in actual use, for example, environmental background information, specifically, for example, a wire, a telegraph pole, a background of the day, a background of the night, and the like, where the target insulator is located. The background information may also be obtained by means of querying or the like. Or, the background of the target insulator in the existing data set is obtained according to the attribute information of the target insulator, and the background is determined as the background information of the defective insulator.
By using attribute information and background information of the target insulator, a preset number of second defective insulator images and corresponding masks thereof are generated by using blend (a piece of open-source cross-platform all-around three-dimensional animation software) software, where the masks show specific positions of insulator defects in the second defective insulator images, and the generated second defective insulator images may be specifically as shown in fig. 3B. The preset number is set by empirical values or historical data. Of course, other images, such as a negative sample, are also included in the first sample set, and the negative sample can be understood as an image without a defective insulator.
In one possible implementation, the defect type of the insulator in the second defective insulator image includes at least one of: insulator breakage and insulator flashover. Of course, all of the defective target insulator images in the first sample set may also include at least one of insulator breakage and insulator flashover, not limited to the second defective insulator image.
In this example, the preset number of second defective insulator images is determined by a 3D simulation technique, so that the first sample set may be expanded, training data may be added, and the detection accuracy of the trained first neural network model may be improved.
In one possible implementation, after acquiring the image to be detected, the method further includes:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise first defect insulator images.
Specifically, the second neural Network uses an algorithm framework of the Fast-RCNN, and the structure of the Fast-RCNN mainly comprises the Fast-RCNN and an RPN (region Proposal Network). Inputting an image to be detected into a second neural network model, inputting the image to be detected into a Res18 structure of a convolutional neural network by an RPN (used for generating a detection frame) network for feature extraction to obtain a feature map, mapping candidate region frames point by point on the feature map through ROI (region of interests), and regressing the parameters of the probability and coordinates of the candidate frames as the foreground through two layers of fully-connected networks.
For ease of fitting network parameters, the coordinate regressors here are not absolute values, but normalized offsets from the anchor coordinates and scale. The parameters of each anchor are x, y, w, h. x, y represent the coordinates of the feature map where the point is mapped onto the original, and w, h represent the width and height of the anchor. Each point on the feature map may generate multiple anchors, depending on the morphology of the target that needs to be detected. In general, the anchor parameters can be generated by clustering the length and width of the target box in the dataset. Finally, the quantities that need to be regressed in the RPN network are Δ x, Δ y, Δ w, Δ h, here all normalized quantities.
In the Fast-RCNN section, some pre-selected boxes with coordinate parameters and foreground probability have been generated in the foregoing method. Screening out a pre-selection frame predicted as a background according to the foreground probability; and sending the remaining pre-selected frames into a convolutional neural network for further adjustment, adding classification, determining whether the insulator exists in the image to be detected according to the final classification, and outputting the position of the insulator in the image to be detected under the condition that the insulator is identified. Since only an insulator, which is one kind of target, needs to be detected, the number of classifications can be set to 2 (background + insulator).
It should be understood that the images in the second sample set for training the second neural network are labeled images, the negative sample is an image not containing an insulator, the positive sample is an image containing an insulator, and since the first defective insulator image in the first sample set is also an image containing an insulator, the first defective insulator image can also be used as training data of the second neural network, and the labeled data of the first defective insulator image is naturally included in the second sample set. During training, the images in the second sample set are detected by the second neural network, and for each image, a number of candidate frames is predicted, the number of candidate frames being generally related to the size of the feature map. In order to stabilize the recall rate of the sample, the size of the feature map is set to be larger, and then the candidate frame is subjected to non-maximum suppression to obtain an optimal target detection frame.
In addition, the performance of the second neural network model needs to be evaluated by adopting the test data set, the test data set with the label is input into the second neural network model for forward processing and post processing, statistical parameters between the detection frame and the real label frame are obtained, and evaluation is carried out according to the statistical parameters.
Referring to fig. 4, fig. 4 is a schematic flowchart illustrating another insulator defect detection method according to an embodiment of the present disclosure. As shown in fig. 4, the method includes:
401. acquiring an image to be detected;
402. under the condition that the image to be detected comprises the insulator, acquiring an insulator image from the image to be detected;
403. determining the insulator image as a target insulator image under the condition that the size of the insulator image is larger than or equal to a preset size;
404. segmenting an insulator image from the image to be detected to obtain a target insulator image;
the insulator image can be understood as an image of an area where the target detection frame is output by the second neural network model, and the insulator image can be understood as a small graph.
The preset size may be set by empirical values or historical data, and one possible preset size is: 300 pixels by 300 pixels. After the insulator image is determined as the target insulator image, the insulator image can be cut from the image to be detected to obtain the cut individual target insulator image.
405. Performing defect detection on the target insulator image to obtain at least one first detection frame;
406. determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
407. and determining that the insulator in the target insulator image has no defect if the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is less than the threshold value.
In this example, whether the insulator has a defect is determined according to the confidence of the at least one first detection frame, so that the accuracy of determining whether the insulator has a defect can be improved.
The steps 401-407 have been described in the embodiment shown in fig. 2, and can achieve the same or similar beneficial effects, which are not described herein again.
Referring to fig. 5, fig. 5 is a schematic flow chart of another insulator defect detection method according to an embodiment of the present disclosure. As shown in fig. 5, the method includes:
501. acquiring a first sample set, wherein the first sample set comprises a first defect insulator image, marking data of the first defect insulator image, a second defect insulator image and a mask of the second defect insulator image;
502. inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection to obtain at least one second detection frame of each image in the first defective insulator image and the second defective insulator image;
503. obtaining a defect detection result according to the confidence of at least one second detection frame;
504. adjusting network parameters of the first neural network according to the defect detection result, the labeling data and the mask code to obtain a first neural network model;
505. acquiring an image to be detected;
506. under the condition that the second neural network model detects that the insulator exists in the image to be detected, a target insulator image is segmented from the image to be detected;
507. and carrying out defect detection on the target insulator image by using the first neural network model so as to determine whether the insulator in the target insulator image has defects.
Whether the insulator has the defect or not can be determined through the defect probability of at least one detection frame when the insulator is detected, and the detection frame is used for detecting the defect. For example, it may be determined that a defect exists when the defect probability is greater than or equal to a certain threshold; when the defect probability is lower than the certain threshold, it is determined that no defect exists.
The steps 501-507 have already been described in the embodiments shown in fig. 2 or fig. 4, and can achieve the same or similar beneficial effects, and are not described herein again.
In accordance with the foregoing embodiments, please refer to fig. 6, fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 6, the electronic device includes a processor, an input device, an output device, and a memory, where the processor, the input device, the output device, and the memory are connected to each other, where the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions, and the program includes instructions for performing the following steps;
acquiring an image to be detected;
under the condition that the image to be detected comprises the insulator, segmenting a target insulator image from the image to be detected;
and detecting the defects of the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects.
The image to be detected is detected, if the insulator exists in the image to be detected, the image to be detected is segmented to obtain a target insulator image, further defect detection is carried out on the insulator in the image based on the target insulator image, whether the insulator has defects or not is determined, and compared with the existing scheme, the method and the device can be used for rapidly positioning the insulator and detecting whether the insulator has defects or not in an artificial mode, and the efficiency of detecting the defects of the insulator is improved.
In one possible implementation, the processor performs the segmentation of the target insulator image from the image to be detected, including:
acquiring an insulator image from an image to be detected;
determining the insulator image as a target insulator image under the condition that the size of the insulator image is larger than or equal to a preset size;
and (4) segmenting the insulator image from the image to be detected to obtain a target insulator image.
In one possible implementation, the processor performs defect detection on the target insulator image to determine whether the insulator in the target insulator image has a defect, including:
performing defect detection on the target insulator image to obtain at least one first detection frame;
and determining whether the insulator in the target insulator image has a defect or not according to the confidence coefficient of the at least one first detection frame.
In one possible implementation, the determining whether the insulator in the target insulator image has a defect according to the confidence of the at least one first detection frame by the processor includes:
determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
and determining that the insulator in the target insulator image has no defect if the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is less than the threshold value.
In one possible implementation, the defect detection of the target insulator image is performed by a first neural network model; the processor is further configured to:
acquiring a false positive rate of the first neural network model and acquiring a recall rate of the first neural network model;
a threshold is determined based on the false positive rate and the recall rate.
In one possible implementation, the processor is further configured to:
determining the insulator in the insulator image as an interference insulator under the condition that the size of the insulator image is smaller than the preset size;
and filtering the image to be detected, and not executing the operation of segmenting the insulator image from the image to be detected.
In one possible implementation, the processor is further configured to:
acquiring a first sample set; the first sample set includes a mask of a first defective insulator image, label data for the first defective insulator image, a second defective insulator image, and a second defective insulator image;
inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection so as to respectively and correspondingly obtain at least one second detection frame;
obtaining a defect detection result according to the confidence of the at least one second detection frame;
and adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
In one possible implementation, the processor is further configured to:
acquiring attribute information and background information of a target insulator;
and generating masks of a preset number of the second defective insulator images and the second defective insulator images by a 3D simulation technology by using the attribute information and the background information.
In one possible implementation, the defect type of the insulator in the second defective insulator image includes at least one of: insulator breakage and insulator flashover.
In one possible implementation, the processor is further configured to:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise first defect insulator images.
The above description has introduced the solution of the embodiment of the present application mainly from the perspective of the method-side implementation process. It is understood that the electronic device comprises corresponding hardware structures and/or software modules for performing the respective functions in order to realize the above-mentioned functions. Those of skill in the art will readily appreciate that the present application is capable of hardware or a combination of hardware and computer software implementing the various illustrative elements and algorithm steps described in connection with the embodiments provided herein. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiment of the present application, the electronic device may be divided into the functional units according to the method example, for example, each functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit. It should be noted that the division of the unit in the embodiment of the present application is schematic, and is only a logic function division, and there may be another division manner in actual implementation.
In accordance with the above, please refer to fig. 7, and fig. 7 is a schematic structural diagram of an insulator defect detecting apparatus according to an embodiment of the present application. As shown in fig. 7, the apparatus includes:
a first acquiring unit 701 configured to acquire an image to be detected;
a second obtaining unit 702, configured to, when the image to be detected includes an insulator, segment a target insulator image from the image to be detected;
a determining unit 703 is configured to perform defect detection on the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects.
In one possible implementation manner, in terms of segmenting the target insulator image from the image to be detected, the first obtaining unit 701 is specifically configured to:
acquiring an insulator image from an image to be detected;
determining the insulator image as a target insulator image under the condition that the size of the insulator image is larger than or equal to a preset size;
and (4) segmenting the insulator image from the image to be detected to obtain a target insulator image.
In one possible implementation manner, in performing defect detection on the target insulator image to determine whether the insulator in the target insulator image has a defect, the second obtaining unit 702 is specifically configured to:
performing defect detection on the target insulator image to obtain at least one first detection frame;
and determining whether the insulator in the target insulator image has a defect or not according to the confidence coefficient of the at least one first detection frame.
In a possible implementation manner, in terms of determining whether an insulator in the target insulator image has a defect according to the confidence of the at least one first detection frame, the second obtaining unit 702 is specifically configured to:
determining that the insulator in the target insulator image has a defect when the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
determining that there is no defect in the insulator in the target insulator image if the maximum confidence in the confidence of the at least one first detection box is less than the threshold.
In one possible implementation manner, as shown in fig. 8, the apparatus further includes a third obtaining unit 704, where the third obtaining unit 704 is configured to:
acquiring a false positive rate of the first neural network model and acquiring a recall rate of the first neural network model;
a threshold is determined based on the false positive rate and the recall rate.
In one possible implementation manner, the second obtaining unit 702 is further configured to:
determining the insulator in the insulator image as an interference insulator under the condition that the size of the insulator image is smaller than the preset size;
and filtering the image to be detected, and not executing the operation of segmenting the insulator image from the image to be detected.
In one possible implementation manner, the third obtaining unit 704 is further configured to:
acquiring a first sample set; the first sample set includes a mask of a first defective insulator image, label data for the first defective insulator image, a second defective insulator image, and a second defective insulator image;
inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection to obtain at least one second detection frame of each image in the first defective insulator image and the second defective insulator image;
obtaining a defect detection result according to the confidence of at least one second detection frame;
and adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
In one possible implementation manner, the third obtaining unit 704 is further configured to:
acquiring attribute information and background information of a target insulator;
and generating a preset number of second defect insulator images and masks of the second defect insulator images by a 3D simulation technology by adopting the attribute information and the background information.
In one possible implementation, the defect type of the insulator in the second defective insulator image includes at least one of: insulator breakage and insulator flashover.
In one possible implementation manner, the first obtaining unit 701 is further configured to:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise first defect insulator images.
Embodiments of the present application also provide a computer-readable storage medium, where the computer-readable storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any one of the insulator defect detection methods described in the above method embodiments.
Embodiments of the present application further provide a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program causes a computer to execute some or all of the steps of any one of the insulator defect detection methods described in the above method embodiments.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some interfaces, devices or units, and may be an electric or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may be implemented in the form of a software program module.
The integrated units, if implemented in the form of software program modules and sold or used as stand-alone products, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be substantially implemented or a part of or all or part of the technical solution contributing to the prior art may be embodied in the form of a software product stored in a memory, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned memory comprises: various media capable of storing program codes, such as a usb disk, a read-only memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and the like.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable memory, which may include: flash memory disks, read-only memory, random access memory, magnetic or optical disks, and the like.
The foregoing detailed description of the embodiments of the present application has been presented to illustrate the principles and implementations of the present application, and the above description of the embodiments is only provided to help understand the method and the core concept of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (13)
1. An insulator defect detection method, characterized in that the method comprises:
acquiring an image to be detected;
under the condition that the image to be detected comprises the insulator, segmenting a target insulator image from the image to be detected;
and detecting the defects of the insulators in the target insulator image to determine whether the insulators in the target insulator image have defects.
2. The method according to claim 1, wherein the segmenting the target insulator image from the image to be detected comprises:
acquiring an insulator image from the image to be detected;
determining the insulator image as the target insulator image when the size of the insulator image is larger than or equal to a preset size;
and segmenting the insulator image from the image to be detected to obtain the target insulator image.
3. The method of claim 2, further comprising:
under the condition that the size of the insulator image is smaller than the preset size, determining that the insulator in the insulator image is an interference insulator;
and filtering the image to be detected, and not executing the operation of segmenting the insulator image from the image to be detected.
4. The method according to any one of claims 1-3, wherein the performing defect detection on the insulators in the target insulator image to determine whether the insulators in the target insulator image are defective comprises:
performing defect detection on the insulators in the target insulator image to obtain at least one first detection frame;
and determining whether the insulator in the target insulator image has a defect or not according to the confidence coefficient of the at least one first detection frame.
5. The method of claim 4, wherein determining whether an insulator in the target insulator image is defective according to the confidence level of the at least one first detection box comprises:
determining that the insulator in the target insulator image has a defect if the maximum confidence coefficient of the confidence coefficients of the at least one first detection frame is greater than or equal to a threshold value;
determining that there is no defect in the insulator in the target insulator image if the maximum confidence in the confidence of the at least one first detection box is less than the threshold.
6. The method of claim 5, wherein the defect detection of the insulator in the target insulator image is performed by a first neural network model; the method further comprises the following steps:
acquiring the false positive rate and the recall rate of the first neural network model for carrying out defect detection on the insulator in the target insulator image;
determining the threshold value according to the false positive rate and the recall rate.
7. The method of claim 6, wherein the first neural network model is trained using the steps of:
acquiring a first sample set; the first sample set comprises a first defective insulator image, label data for the first defective insulator image, a second defective insulator image, and a mask for the second defective insulator image;
inputting the first defective insulator image and the second defective insulator image into a first neural network for defect detection so as to respectively and correspondingly obtain at least one second detection frame;
obtaining a defect detection result according to the confidence of the at least one second detection frame;
and adjusting the network parameters of the first neural network according to the defect detection result, the labeling data and the mask.
8. The method of claim 7, wherein prior to obtaining the first set of samples, the method further comprises:
acquiring attribute information and background information of a target insulator;
and generating masks of a preset number of the second defective insulator images and the second defective insulator images by a 3D simulation technology by using the attribute information and the background information.
9. The method of claim 7 or 8, wherein the defect type of the insulator in the second defective insulator image comprises at least one of: insulator breakage and insulator flashover.
10. The method according to claim 7 or 8, characterized in that after acquiring the image to be detected, the method further comprises:
identifying the image to be detected by adopting a trained second neural network model so as to determine whether the image to be detected comprises an insulator; the second neural network model is obtained by training a second neural network by adopting a second sample set, the second sample set comprises images not containing insulators and images containing insulators, and the images containing insulators comprise the first defect insulator image.
11. An insulator defect detection apparatus, the apparatus comprising:
the first acquisition unit is used for acquiring an image to be detected;
the second acquisition unit is used for segmenting a target insulator image from the image to be detected under the condition that the image to be detected comprises the insulator;
and the determining unit is used for detecting the defects of the insulators in the target insulator image so as to determine whether the insulators in the target insulator image have defects.
12. An electronic device comprising a processor, an input device, an output device, and a memory, the processor, the input device, the output device, and the memory being interconnected, wherein the memory is configured to store a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of any of claims 1-10.
13. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program comprising program instructions that, when executed by a processor, cause the processor to carry out the method according to any one of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011610362.4A CN112634254A (en) | 2020-12-29 | 2020-12-29 | Insulator defect detection method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011610362.4A CN112634254A (en) | 2020-12-29 | 2020-12-29 | Insulator defect detection method and related device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112634254A true CN112634254A (en) | 2021-04-09 |
Family
ID=75286802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011610362.4A Pending CN112634254A (en) | 2020-12-29 | 2020-12-29 | Insulator defect detection method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112634254A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113192071A (en) * | 2021-06-29 | 2021-07-30 | 南方电网数字电网研究院有限公司 | Insulator binding wire missing identification method and device and computer equipment |
| CN113506293A (en) * | 2021-09-08 | 2021-10-15 | 成都数联云算科技有限公司 | Image processing method, device, equipment and storage medium |
| CN113538412A (en) * | 2021-08-06 | 2021-10-22 | 广东电网有限责任公司 | Insulator defect detection method and device for aerial image |
| CN113538411A (en) * | 2021-08-06 | 2021-10-22 | 广东电网有限责任公司 | Insulator defect detection method and device |
| WO2024021662A1 (en) * | 2022-07-29 | 2024-02-01 | 宁德时代新能源科技股份有限公司 | Method and apparatus for detecting defect of battery cell surface |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615611A (en) * | 2018-11-19 | 2019-04-12 | 国家电网有限公司 | A kind of insulator self-destruction defect inspection method based on inspection image |
| CN109801284A (en) * | 2019-01-25 | 2019-05-24 | 华中科技大学 | A kind of high iron catenary insulator breakdown detection method based on deep learning |
| CN110992314A (en) * | 2019-11-15 | 2020-04-10 | 广东华路交通科技有限公司 | Pavement defect detection method and device and storage medium |
| CN111402215A (en) * | 2020-03-07 | 2020-07-10 | 西南交通大学 | Contact net insulator state detection method based on robust principal component analysis method |
| CN111583198A (en) * | 2020-04-23 | 2020-08-25 | 浙江大学 | Insulator picture defect detection method combining FasterR-CNN + ResNet101+ FPN |
| CN111784692A (en) * | 2020-08-11 | 2020-10-16 | 国网内蒙古东部电力有限公司 | Method and device for detecting insulator defects in power system and electronic equipment |
| CN111814867A (en) * | 2020-07-03 | 2020-10-23 | 浙江大华技术股份有限公司 | Defect detection model training method, defect detection method and related device |
| CN111814850A (en) * | 2020-06-22 | 2020-10-23 | 浙江大华技术股份有限公司 | Defect detection model training method, defect detection method and related device |
| CN112001902A (en) * | 2020-08-19 | 2020-11-27 | 上海商汤智能科技有限公司 | Defect detection method and related device, equipment and storage medium |
-
2020
- 2020-12-29 CN CN202011610362.4A patent/CN112634254A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615611A (en) * | 2018-11-19 | 2019-04-12 | 国家电网有限公司 | A kind of insulator self-destruction defect inspection method based on inspection image |
| CN109801284A (en) * | 2019-01-25 | 2019-05-24 | 华中科技大学 | A kind of high iron catenary insulator breakdown detection method based on deep learning |
| CN110992314A (en) * | 2019-11-15 | 2020-04-10 | 广东华路交通科技有限公司 | Pavement defect detection method and device and storage medium |
| CN111402215A (en) * | 2020-03-07 | 2020-07-10 | 西南交通大学 | Contact net insulator state detection method based on robust principal component analysis method |
| CN111583198A (en) * | 2020-04-23 | 2020-08-25 | 浙江大学 | Insulator picture defect detection method combining FasterR-CNN + ResNet101+ FPN |
| CN111814850A (en) * | 2020-06-22 | 2020-10-23 | 浙江大华技术股份有限公司 | Defect detection model training method, defect detection method and related device |
| CN111814867A (en) * | 2020-07-03 | 2020-10-23 | 浙江大华技术股份有限公司 | Defect detection model training method, defect detection method and related device |
| CN111784692A (en) * | 2020-08-11 | 2020-10-16 | 国网内蒙古东部电力有限公司 | Method and device for detecting insulator defects in power system and electronic equipment |
| CN112001902A (en) * | 2020-08-19 | 2020-11-27 | 上海商汤智能科技有限公司 | Defect detection method and related device, equipment and storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113192071A (en) * | 2021-06-29 | 2021-07-30 | 南方电网数字电网研究院有限公司 | Insulator binding wire missing identification method and device and computer equipment |
| CN113538412A (en) * | 2021-08-06 | 2021-10-22 | 广东电网有限责任公司 | Insulator defect detection method and device for aerial image |
| CN113538411A (en) * | 2021-08-06 | 2021-10-22 | 广东电网有限责任公司 | Insulator defect detection method and device |
| CN113506293A (en) * | 2021-09-08 | 2021-10-15 | 成都数联云算科技有限公司 | Image processing method, device, equipment and storage medium |
| WO2024021662A1 (en) * | 2022-07-29 | 2024-02-01 | 宁德时代新能源科技股份有限公司 | Method and apparatus for detecting defect of battery cell surface |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112634254A (en) | Insulator defect detection method and related device | |
| CN111179249A (en) | Power equipment detection method and device based on deep convolutional neural network | |
| CN113177469B (en) | Training method and device of human attribute detection model, electronic equipment and medium | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN112733690B (en) | High-altitude parabolic detection method and device and electronic equipment | |
| CN110647822A (en) | High-altitude parabolic behavior identification method and device, storage medium and electronic equipment | |
| CN112200081A (en) | Abnormal behavior identification method and device, electronic equipment and storage medium | |
| CN110991448A (en) | Text detection method and device for nameplate image of power equipment | |
| CN111723773B (en) | Method and device for detecting carryover, electronic equipment and readable storage medium | |
| CN112967248B (en) | Method, apparatus, medium and program product for generating defect image samples | |
| CN116168351B (en) | Inspection method and device for power equipment | |
| CN113870254B (en) | Target object detection method and device, electronic equipment and storage medium | |
| CN113255590A (en) | Defect detection model training method, defect detection method, device and system | |
| CN116647644B (en) | Campus interactive monitoring method and system based on digital twin technology | |
| CN112819796A (en) | Tobacco shred foreign matter identification method and equipment | |
| CN113516146A (en) | Data classification method, computer and readable storage medium | |
| CN112330597A (en) | Image difference detection method and device and computer equipment | |
| CN109740527B (en) | Image processing method in video frame | |
| CN114119528A (en) | Defect detection method and device for distribution line insulator | |
| CN114022811A (en) | Water surface floater monitoring method and system based on continuous learning | |
| CN113065533A (en) | Feature extraction model generation method and device, electronic equipment and storage medium | |
| CN106651803B (en) | Method and device for identifying house type data | |
| CN111444864B (en) | Method and device for detecting abrasion degree of target object | |
| CN113505643A (en) | Violation target detection method and related device | |
| CN112669302A (en) | Dropper defect detection method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210409 |